One thing I like about the study of economics is that it fosters compassion. When part of your job is to predict human behavior, you quickly learn the value of understanding other people’s problems. When the other part of your job is ferreting out the unseen global consequences of our choices, you’ve taken the first step toward caring about those consequences.
For example: Suppose a guy with no health insurance and no assets shows up at a hospital emergency room with an urgent life-threatening condition. Should you let him die? Ordinary compassion says no. The heightened compassion of the economist says, at the very least, maybe.
First, a policy of providing emergency health care to everyone is pretty much the same thing as a policy of providing emergency health insurance to everyone. It was specified here that this was a guy who didn’t want health insurance. So let’s recognize for starters that such a policy runs counter to — I am tempted to say runs roughshod over — the guy’s own revealed preference. It’s an odd sort of compassion that forces people to buy things they don’t want.
Now you might object that nobody’s forcing this guy to buy emergency health care; we’re trying to give him emergency health care. Not so fast. Here’s the first place where a little economic training goes to hone one’s sense of compassion: The emergency health insurance we’re foisting on this guy has a cost. We can spend that money on emergency rooms or we can spend it on a myriad of other things the guy might prefer. How is it compassionate to give him one thing when he prefers another?
This is particularly true if the guy happens to be very poor. Poor people have a lot of problems, and emergency health care is only one of them. They need better education, they need better transportation, and they need a little help buying groceries.
There is room for lots of debate and lots of disagreement about how much we as a society should be spending to help poor people. That’s not the issue here. The issue here is: Given that you’ve decided to spend an extra such-and-such many dollars a year helping poor people, why would you spend it in this particular way rather than one of the many other ways they could use it? For God’s sake, why not at least ask them if they’d rather have the cash?
There are many good economic arguments for subsidizing health care (there are also many good counter-arguments). That’s yet another important debate that’s largely off-topic here. I want to focus attention on the narrower question of what compassion demands.
Here’s my answer: If your compassion is constrained, blind or posturing, you’ll say “Of course we should save lives at the emergency room”. If your compassion is broad, perceptive and genuine — if you grasp and care about the underlying trade-offs — and if you believe you’re being baited by a constrained, blind, posturing journalist — you might very well burst forth with a (com)passionate “Let him die”.
There’s yet another facet to this compassion business, which I’ll mention only briefly because Robin Hanson got there first: Of the many commentators who have jumped in to decry the “lack of compassion” among the debate audience, I’ll wager that nearly all are perfectly comfortable with the American government’s current policy of providing American style health care only to Americans, as opposed to, say, Kazakhs. In other words, they too are perfectly fine with a policy of “let him die”, as long as he dies far enough away. At this point, we’re not arguing about principle; we’re arguing about where to draw an arbitrary line. Once you’ve admitted that there are limits to compassion, it’s fair to ask: Why is it okay for Americans to ignore the plight of Kazakhs, but not okay for Texans to ignore the plight of New Yorkers, or for rural Pennsylvanians to ignore the plight of Philadephians? Maybe there are answers to those questions, but if so, those answers are surely not to be found in the mindless cry of “compassion”.
There are deep and troubling issues here that deserve to be addressed, and economics has a lot to contribute to that discussion. That’s why it’s so very disappointing to see Paul Krugman, yet again, using his platform in the New York Times to bray with the yahoos instead of calling attention to deep and difficult trade-offs. None of us have all the answers, but that’s no excuse for pretending there are no hard questions.
There’s another side to compassion, and that’s a willingness to admit that your adversaries might occasionally have legitimate or even lofty motives, and that their positions are sometimes worthy to be engaged, not simply pilloried. In pursuit of a narrow political agenda, Krugman betrays a failure of compassion not just toward those he disagrees with, but, more importantly, to the many Americans (including many of the desperately poor Americans for whom he so often and so self-righteously claims to speak) who stand to benefit from the wiser policies that just might follow from the more informed public discussion to which I know Paul Krugman could contribute brilliantly if he so chose. Instead we have the sad irony of an embittered polemicist wasting his talent by spewing vitriol against everyone who dares to struggle with the difficult policy choices that he has decided are beneath his notice — all in the name of compassion.
Note: You’ll find more on the economics of compassion in Chapter 15 of my book More Sex is Safer Sex.
In your example of the guy with no health insurance having a life threatening illness, you clearly state that he has showed up to an emergency room. To me this strongly suggests that he would like to recieve medical treatment rather than help buying next weeks groceries.
It was also specified that he showed up at the emergency room.
Now the response to that might be that you should inform him that you don’t have the money to help him because, in light of his revealed preference from groceries over insurance, the money was given to a local food bank instead. I get that that’s the point being made here. But can’t we also assume that there will always be a certain number of poor uninsured people who will, despite this, show up at emergency rooms with the new revealed preference that they would rather you save their life than have as many groceries in the future? Does taking that probability into account justify some charitable resource allocation deviating from revealed time preferences?
I’m 99% certain there’s a serious flaw in your logic of compassion here, but don’t have time to explore it now. It’s also sad that you are so vitriolic when you accuse Krugman of being vitriolic. :-/
It wasn’t specified that he didn’t want health insurance, merely that he didn’t have any. There are lots of things that people want, or even need, which sometimes they cannot afford.
And to suggest that perhaps he should buy health insurance by spending less on groceries… for many uninsured people, that would likely mean getting insurance to pay for medical treatment for the starvation they just chose because of their very limited resources.
Neverfox:
But can’t we also assume that there will always be a certain number of poor uninsured people who will, despite this, show up at emergency rooms with the new revealed preference that they would rather you save their life than have as many groceries in the future?
Yes. We can also assume there will always be a certain number of poor people who will really appreciate some help at that food bank. I object to ignoring those people in the calculus of compassion.
Does taking that probability into account justify some charitable resource allocation deviating from revealed time preferences?
We can fully take that probability into account, at least in principle, by asking poor people which they would prefer — help at the emergency room or help at the food bank. They are presumably aware that there’s some probability they’ll show up at the emergency room, and will account for that in their answers.
Maybe they’d prefer the help at the emergency room. Maybe they’d prefer the help at the food bank. I don’t see any sense in which assuming the former is more compassionate than assuming the latter.
Matthew:
To me this strongly suggests that he would like to recieve medical treatment rather than help buying next weeks groceries.
I’m quite sure of that. I’m also quite sure there are many others who would prefer help buying next week’s groceries to an assurance that they’ll get medical treatment if they need it. The whole point is that full-blooded compassion accounts for those people also.
Andrew:
It wasn’t specified that he didn’t want health insurance, merely that he didn’t have any.
If he doesn’t know his own preferences, why would you expect a policymaker to know them any better?
It could be rational to support free emergency room treatment for the poor (as opposed to letting the poor choose the resource allocation) if you can show that free medical care leads to less of a “free rider” problem than other forms of assistance.
If you hand out free apples, plausibly a lot of people will choose not to earn the money that they could have earned to by apples, and just take the free ones. But if you hand out free emergency room treatment, far fewer people are going to go out and get life-threatening injuries just to qualify for the freebies.
There are a few problems with this argument. First and foremost, there is a difference between saying “we have considered the policy implications and there may be reasons why we should spend our limited resources on things other than providing free medical care” and shouting out “let him die!” I have some serious doubts that the audience member at the Republican debate who shouted this out did so after making this sort of policy consideration.
Second, there is arguably a problem relying on people’s choices between short term benefits versus preparing for a long-term unlikely contingency. There is good reason to believe that people have difficulty evaluating (and therefore valuing) the risk that they might come down with an expensive, life threatening disease at age 30. Since 30 year-olds rarely get such diseases, and people are bad at properly comprehending the risk of low likelihood outcomes, most people set the cost of such a disease at 0% instead of at say, 1%. While there is a very small absolute difference between these risks, the difference between these two views has a significant impact on how you decide how to allocate resources. The expected value of protecting against a 1% chance of a disease with a $1,000,000 cost is $10,000. The expected value of protecting against a disease with a 0% chance and a $1,000,000 cost is $0.
Finally, part of the compassion issue is that many people have looked at the US’s resources and decided that, in fact, there are sufficient resources to both ensure that the poor have food and ensure that they have access to medical care. The decision to “let him die” is not a decision to allocate money to a food bank instead, but a decision to allocate that money to providing a tax deduction for corporate jets, and it is unlikely that the dying man has made that revealed preference (unless he was an audience member at a Republican Debate). We don’t try to provide medial care to the whole world because it is less clear that we have the resources to do that (although we do try to provide a certain level of medical care through US aid efforts).
It seems to me that at least some of the commenters are imagining a different scenario to the one Steve is talking about.
If I am right, Steve is imagining a situation where you say to a poor person on day 1: “Would you like us to regularly divert these resources to you to buy food, or would you rather we diverted a lesser amount to buy food each week and instead buy you some health insurance to cover the eventuality that you get sick”.
Whereas some commenters are imagining a situation where each week we ask the question “Do you want your help this week in the form of food or less food plus medical coverage.”
Obviously, in the latter scenario people will aceept the food until they are ill, at which point they will suddenly realise the need to have medical coverage. Whereas Steve is considering the more difficult former case, where a poor person had to look beyond their immediate needs, and has chosen an option which he subsequently might come to regret.
Replace “health insurance” by “vaccine”.
You ask poor people, we have $X to spend on your welfare. Would you rather have food, or a vaccine against these 100 untreatable diseases? They choose food.
Then, one of them shows up at the emergency room with one of the diseases. “Sorry,” we say, truthfully, “there’s nothing we can do for you. The disease is untreatable.”
Does that mean that we were lacking compassion in not buying the poor people the vaccine? No, because we gave them food instead, at their own request.
I think most people, including commenters here, would not have a big problem with that argument.
What’s the difference between this problem and Steve’s? In this problem, there’s nothing we can do once the victim is identified, no matter how much we’re willing to spend. In the other problem, we have to say no to a very specific, desperate victim.
That’s what makes Steve’s problem so difficult. It forces us to think about looking at a specific person in the eye, and saying, “no, you’re going to die because we don’t want to spend an extra $X to save you, especially after we spent $Y on food for you at your request instead.”
What I think Steve is saying, is, how is that so much worse than saying, “you’re going to die because, earlier, we didn’t want to spend an extra $X to buy you a vaccine, especially after we spent $Y on food for you at your request instead?”
I think that morally, it’s NOT worse. It just forces us to FEEL worse — much, much, worse. Well, me, anyway. :)
The thing that is unique about health insurance is the moral hazard problem. It mightake sense to force poor people to buy insurance if none would be able to buy without most in their “group” (the group of poor people) buying in when they’re healthy. So, the choice effectively could become relatively cheap insurance for all, or only very expensive insurance for the few in the poor group.
This is a question related to ethics and moral, a filed that is not covered at all by economics.
.
Filling a paper with plenty of misplaced mathematics does not turn a ideology into science.
Steve has missed the distinction between expressed and revealed DEFINITIONS. Whatever Krugman SAYS he means by compassion he FUNCTIONALLY means “what I think and republicans don’t”.
Moreover, it is not very smart to suppose that the preferences does not change with time. The hypothesis of an static preference is nothing more than bad science, and as we are talking about economics, bad human science.
I saw two additional arbitrary lines in Blitzer’s hypothetical. For one thing, there is nothing special about $1 million. If it cost $10 million or a billion to fix the young man’s condition, most people would probably say “No, it’s not worth spending that much to cure him.” If there is a moral principle here, it’s a strange one, because it apparently only applies to a million dollars and not much larger amounts.
For another thing, there is something arbitrary about discussing a cure that works 100% of the time. What if the million dollar cure had only a 50% chance of curing him? How about a 5% chance? Is it still worth spending a million dollars to save him? I’m guessing most people would say “no” if you set the probability of recovery low enough.
What moral principle has such arbitrary cut-offs?
I love it how there is a category topic called Paul Krugman.
This is the same issue as in Panglossian economics, except with an increase in spending rather than a cut. Except Steve is consistent on it and Krugman has changed sides since it suits his agenda.
This somewhat reminds me of the Dennis Moore Monty Python skit. After robbing from the rich and giving to the poor, so that now the rich are poor and the poor are rich, Moore declares, “This redistribution of wealth is trickier than I thought!”
Steve’s argument explicitly depends on there being a fixed pot of resources which one can devote to, say, food or emergency health care. But it seems to me that Krugman is calling for more total allocation of resources to the (American) poor, in part in the form of emergency health care. So the relevant tradeoff is tax rates versus emergency health care, or the like.
I have a lot of thoughts on this, but one of which is that if I get into an accident or have a sudden health problem and am rushed to the emergency room without proof that I have health insurance (which I do), I would like for them to treat me anyway rather than take time trying to figure out if I am able to pay them.
@vinc,
I think Steve’s argument is that those tax rates (whatever they may be, that is not the argument here and plenty of people can disagree where that specific line should be set) can go towards other goods and services that the poor may deem more valuable – one being the food bank. When people, like political Paul Krugman on his NYT blog (he does good economics outside of the blog, from what I can tell), blatantly disregard these trade-offs for the quick exit of saying, “how dare they” in the name of compassion then something is amiss here.
I think Steve’s argument is simpler: Krugman is ignoring complexity in order to pontificate; that is neither compassionate nor useful. Compassion should extend beyond JUST the man in the ER.
Minor point first:
Andrew: “It wasn’t specified that he didn’t want health insurance, merely that he didn’t have any.
Steve: If he doesn’t know his own preferences, why would you expect a policymaker to know them any better?”
So it was not specified that he didn’t want health insurance, only that he didn’t want health insurance as much as food. I am sure if offered free health insurance, he would accept, indicating that probably he did want health insurance. Alternatively, keep offering him food and health insurance, and when he is really full perhaps he will take the insurance.
More substantively, these are very interesting issues, and deserve careful thought.
My view is that everyone in a country as rich as most western countries should have both food and emergency medical care. You say “Given that you’ve decided to spend an extra such-and-such many dollars a year helping poor people” well, that amount in my view should be sufficient to provide healthcare and food and shelter. You could say this was dodging the question, and you would be right, but I further develop the idea below. How do I know the recipients wouldn’t prefer transport vouchers, or cinema tickets?
The crux of this one comes down to the abilty of people to make rational, consistent choices. It may be that asking people when they are hungry whether they prefer food to emergency room support would not elicit the response that gives the greatest “utility”.
We all have instincts, and we all have intellect. Our instincts lead directly to drives over which we have little conscious control. These evolved before intellect. Most people would probably say that maximim utility – perhaps we could say happiness- derives to a large extent from the intellect. We like to think that we can get more out of life than a dog or cat, although these animals are capable of contentment, possibly happiness (in my view).
This leads us to a conflict between our drives and our intellect. Our drives tell us to take a particular course of action to fulfill our instincts. Sometimes this will be at odds with what our intellect tells us is the action that will lead to maximim utility. Sometimes our intellect wins, sometimes not. I think it widely agreed that happiness is not likely to follow unless the most basic drives are satisfied.
Does this tell us anything about policy choices? Well it does if there is a consistent, predictable mis-match between drive and intellect. I suggest that this is the case if someone is hungry – the drive to eat will take precedence over any long-term planning. This is necessary to ensure survival. I suspect that many here would agree, that if someone is actually hungry, or in pain, they are not going to make good choices.
There is a well known heirarchy of needs. At the bottom comes food, shelter etc. I would add health. If these basic needs are not fulfilled, then humans are not capable of making rational choices.
So I believe that there is a minimum standard of living below which it does not make much sense to talk about economic theory. This should compel the compassionate to provide at least these basic needs, before any talk of choice is discussed.
I have assumed that the wealth of the country is sufficient to provide this – such as the USA.
The above discussion does not solve the problem, because once these basic needs are met there will be further choices to be made. Perhaps we could issue travel permits for job interviews, or we could issue drugs and alcohol vouchers. Should we let the poor choose? Again, there is good reason to believe that poverty can lead people to make choices that would not lead to long-term utility maximisation. I believe that there are predictable mis-matches, and hence we should not necessarily do what the recipients ask. The reason that a regulator may know better than the poor person is because they have access to a large body of research, which may be able to inform about consistent mis-matches between choices and utility.
I think the above is pretty sound when it comes to the basic provision of food, shelter etc. When it comes to informing higher level choices, then it is incumbent on the policy maker to demonstrate how his access to the research body informs him to override the choices of individuals. Evidence based policy, basically. I fear this is sadly lacking in favor of ideology based policy.
Apologies this is so long, I don’t have time to make it any shorter.
Harold:
This should compel the compassionate to provide at least these basic needs, before any talk of choice is discussed.
Why would compassion compel you to provide what you call “basic needs” for Americans, while ignoring the basic needs of sub-saharan Africans?
Why might not the same compassion compel you to provide basic needs for, say, fellow residents of your home state but not of other states? Or of your family but not of other families?
As I said in the post, there might be answers to these questions, but I do not think it is helpful simply to claim that we can settle these hard questions with an appeal to the abstract principle of compassion.
Several other commentators already beat me to the point I’m about to readdress, but I’d like to reaffirm it anyway.
Steve’s argument seems to be the equivalent of asking an audience of people who’d they’d like me to save first out of two drowning children in the distance, child A or child B. Now, assuming I’m the only one who is able to swim (or at least fast enough to save one), you’d think that given a limited amount of time the audience would choose to have me save at least one. But what if I’m a really fast, efficient, and strong swimmer who is able to save both. It’d seem a little foolish to ask them to decide when they would simply ask “Can’t you save both?”
And that’s the issue here with food and healthcare for the poor in the USA. You can easily do both. We’re not a developing country that has to face daily trade-offs between very limited resources, so why act like we are?
I don’t. That’s why I’m an anarchist. But this doesn’t have to be about policymakers and top-down distribution. I might need to make a personal decision about which charity programs will get my money. It’s not realistic to tell me to ask every poor person I hope to help.
My point is that if we are talking about helping the poor with extra resources, what we’re hopefully trying to achieve is a lump of resources that, when spread around, make them (at least functionally) not poor anymore. We give money because we want to hold their heads above the water, not keep them at or below it. So the allocation preferences shouldn’t reflect the state of being poor, but the state of being slightly (or maybe even, much) better off than poor. That would seem to make the revealed preference data gathered before the point that they have been raised out of that irrelevant at best. That’s like taking my preferences when I have $5 in my wallet and, from that, trying to say what I’d do if I had $100. I’m not saying that helps us get closer to an answer, but it tells us that our answer is a priori likely to be wrong if we use that data. That hardly seems like something to hang our compassionate hat on.
Thus far, I’ve been taking revealed preference to be valid for the sake of argument. Of course, revealed preference is bunk and that should probably be addressed at some point in this discussion.
@NeverFox: I express a preference for losing weight. But I skip the gym too often to actually lose weight, I go the bakery instead. My expressed preference differs from my revealed preference. Explain why that is “bunk”.
You ask me who I will vote for. I say Obama but just because I swim in a liberal sea; I end up voting for McCain. My revealed preference differs from my expressed prefernce — in this case due to duplicitly. Explain what that is “bunk”.
Don’t give me a story about reconstructing my indifference curves because my examples do not rely on such curves even existing. It merely illustrates that what I say even about my preferences can be wrong. I am asking you to debunk THAT.
I don’t have to debunk that because you’re not talking about the neoclassical concept of revealed preference, but rather demonstrated preference, something I don’t have any problem with. Maybe Steve intended to talk about the latter, but let’s not pretend that the term “revealed preference” isn’t already defined in the economic literature as having to do with order preferences and indifference curves.
IMO, the discussion is a moot one. Given the current healthcare insurance mess in the US, there is no reason to make the “life or death” choice based on cost. Implementing a UHC (or SP/UHC) system reduces costs by 40% to 50+% on an annual basis if costs (ppp basis) were to merely *match* those of France (rated #1 in world for healthcare by OECD in 2007). Thus, as *everyone* is covered under a UHC system, everyone pays based on what they can afford.
The simple fact the individual *chose* to go to an ER means ther person was choosing to game the system. Under EMTALA, the person KNEW if he/she became seriously ill, the ER was the place to go if there was nowhere else to turn. And that is what conservatives choose to cite as them providing “public healthcare”. However, what is omitted is the fact most ER visits (think of those recent deaths due to *dental* infections) can be totally eliminated by having proper medical and dental care in the first place.
Plus, EMTALA “sort of” made a govt public policy statement: Life is more valuable than money. Now, if that policy changes, then violence becomes an established part of US life–because killing someone is no longer a *criminal* offense, it is merely a matter of paying off the family for their money loss when someone is killed (accidentally OR deliberately). In order to prevent poor people from going on killing sprees, the govt would have to reinstitute debtor’s prisons–no getting out until debts are paid (otherwise, no way to stop someone from killing a lot of people and claiming “poverty”). It would ALSO mean the rich could afford to pay for–and hunt–people. Now *there* is a proper business venture with Dick Cheney as CEO….
Then there is Johathan Swift and “A Modest Proposal”…. Two (or more) problems fixed with one solution.
A lot more can be done once there is a *maximum* price on a life. Use your imagination.
I’m curious – when Landsburg spends money, does he regularly consult strangers to see what they would prefer?
I disagree with Landsburg’s fundamental premise that compassion generally motivates people simply to transfer wealth to the poor. Rather, I think compassion causes people to do unto others as they wish others would do unto them – NOT to do unto others as others would prefer. Compassion is not some magical, mystical impulse, distinct from other impulses; compassion just one more normal, conventional impulse. As with any other impulse, a person motivated by compassion will do things that maximize the utility of the PERSON WHO IS MOTIVATED BY COMPASSION – not maximize the utility of the person who is the object of the compassion.
Thus, I support health care for the poor because I prefer not seeing sick people. The poor prefer something else? Fine; let them pay for the something else with their own resources. But if I’m paying the piper, I’m calling the tune.
Now, I also suspect that lack of regular health care for the poor also results in reduced efficiencies for our health care system overall, and that negative externalities arise from private health care decisions; I suspect that these dynamics swamp the inefficiencies that result from subsidizing health care services. (I believe I share Landsburg’s concerns about the licensing of health care professionals, but that’s another thread.) But that may simply be icing on the cake. I like what I like.
Consider various forms of disaster relief, including FEMA intervention and farm subsidies. As an initial proposition, I could argue that government intervention reduces people’s incentives to protect themselves. But I’d have to be a pretty slow learner not to discover that during an emergency, compassionate impulses will overwhelm this general principle of non-intervention. Consequently the second-best remedy may be to design efficient mechanisms for intervening, and subsidizing these mechanisms by compelling people to buy various forms of insurance – flood insurance, crop insurance, health insurance, etc.
It’s all well and good for the libertarian to say, “Let people run the risks they choose; it’s not your concern if they do.” But this is the difference between philosophy and social science: overwhelming evidence indicates that, wisely or not, people ARE concerned. My utility is affected by my perception of your utility. Your reckless disregard for your own safety affects me.
And, in contrast, the burdens you bear by being compelled to buy insurance do not affect my utility function as much. What can I tell you? I’m shallow – and I suspect I’m representative of the public at large.
To misappropriate Douglas Adams, libertarians espouse optimal policies for a race of beings who are almost, but not quite, entirely unlike humans. Once we inject this sticky concept of compassion into the analysis – the idea that I’m trying to maximize MY utility which is, in part, a function of MY perception of YOUR utility — the calculus gets much messier.
If we pour enough money into health care such that the tailor decides he’s better off answering phones at the local hospital than sewing, who’s going to make the little hats that nurses wear? (Do they still wear those hats?)
Human want is infinite. Resources are scarce. Waving your hands and saying “We are a rich county, we can do everything everywhere all the time!” is just naive.
I know this is off topic, but my guess is that if we didn’t have much gov’t intervention in the health insurance and medical care markets, this wouldn’t even be a discussion.
True health insurance (that is, something that pays for big, unexpected medical expenses) might come in the form of a bene that you get when you use a credit card. Sort of like how you get travel and accident insurance when you buy plane tickets with certain credit cards.
And the routine stuff would be rather inexpensive. ‘We’ might decide to target some help for some folks, like with food stamps, but we wouldn’t make it the rule.
Apply a lot of the same interventions that we have in health insurance/medical care to the shoe market and we’d be discussing the compassion about the shoeless. And there would be many more shoeless than we have today.
“Why would compassion compel you to provide what you call “basic needs” for Americans, while ignoring the basic needs of sub-saharan Africans?”
This is indeed a very good point, and a very difficult question. My earlier response avoided this particular issue as it was long and rambling enough, but this issue is related.
I argued that individual’s preferences may not lead to their maximum utility in part because of a split between innate drives and intellectual appreciation of happiness. Also that this could be predictable and consistent enough to influence policy that overrides individual’s preferences, expressed or revealed.
Lets define compassion as the desire to help others less fortunate at some cost to yourself. How could such a behaviour evolve? It is through the mechanism of group selection, where the group, and hence individuals in the group, survives better if members of the group exhibit such behaviour. This has been dressed up as “kin selection”, but it boils down to much the same thing. For this behaviour to evolve, a strong sense of the group is needed. This motivates you to help other members of the group, but not “outsiders”. Thus for altruism to evolve, something akin to patriotism is needed. We could call this our innate compassion.
On top of this we can layer an intellectual appreciation of compassion. If this is truly altruistic, then there is no reason to discriminate among those you wish to help. It is hard to define a member of your own group as intrinsically more valuable than an outsider, if you are attempting to be objective. We have again a conflict between our insticts and our intellect. The economic argument is pure intellect, as far as I can see. This is why it is in conflict with our actual preferences, which are very much influenced by our innate drives.
So how do we resolve this? We can see why the policies we have exist – the country is the tribe, and therefore altruism within the tribe is compliant with our insticts. Altruism outside the tribe is not. To put it another way, who’s utility are we trying to maximise? Most policy concerns itself only with the in-group. We could say, that’s fine, that is what we care about. But to do so we should recognise the lack of intellectual rigour in that approach. To show similar compassion for the poor of Africa requires us to leave instinct behind, and become truly reasoning beings.
To start by quoting Mike H above: It’s also sad that you are so vitriolic when you accuse Krugman of being vitriolic. :-/
You state that the example is a man with no health insurance and no assets, and later you state that he made a choice to not want health insurance. As far as I can tell, one does not logically follow from the other. If he has no assets, is it not then also true that he cannot afford health insurance? You may say that he made a choice to spend his meager earnings, if there were any one food, which would probably change the example little, but to say that he didn’t want health insurance is not the same thing as not having it. He may have wanted health insurance, he may just not have wanted it as badly as whatever other necessities he presumably spent that money on.
Also, in terms of compassion, your definition of compassion seems to be a compassion for allowing people to make their own choices and allowing them to live with the responsibility of compassion. If you would like to argue from that point, fine, but that is NOT the definition of compassion. The definition of compassion is: a feeling of deep sympathy and sorrow for another who is stricken by misfortune, accompanied by a strong desire to alleviate the suffering. Based on this definition, as well as the colloquial definition that I believe most people (although not you apparently) would accept, compassion is a feeling for AN INDIVIDUAL, not for the nameless, faceless world at large. If you want to argue that letting the man die is better for the world at large, fine, but I object to your argument that this is compassion as most people would define it. Make up a new word if you must, but compassion is an EMOTION, and emotions do not always follow the law of logic.
Also, it seems a strange sort of compassion for someone when they are forced to buy health insurance, but not when they are dying in a hospital (or, as the case may be, on the street) seems to me the complete opposite of compassion. I surely hope that the greatest suffering in your life is being forced to buy something you don’t want, particularly when that something is designed to help limit a far greater suffering later in life. You may not like buying insurance (although it would shock me if you didn’t have it), but I have little compassion for you if that is the absolute worst thing that you will face in your life. I guess you can just call me heartless.
You assume that “They are presumably aware that there’s some probability they’ll show up at the emergency room, and will account for that in their answers,” and this is my biggest problem with the field of economics that attempts to analyze society. You seem to assume that all people think as logically as you do. Most do not. Most people might realize in the back of their mind that they could always show up at the emergency room, but if you asked someone what they would like X amount of money spent on, they will probably think more immediately, not in the indefinite future. You could give a multiple choice question, but then you would always undoubtedly leave a choice out.
I realize the last bit maybe understood as a bit off topic, but it is there nevertheless. To be more on point, you state that his revealed preferences are to pay for other things over health care. Could it not be argued that he would like to stay alive and pay for whatever the cost, assuming he was able? His preference two weeks earlier might have been food, but when he got into a car accident, for example, he probably preferred the emergency medical procedures necessary to keep him from death. As for the man that could not afford the procedure, I’m sure you see it as compassionate to have just let him die out on the street. I’m also sure that you would have been among those most vehemently opposed to giving any aid whatsoever to the famine Irish. This was first paid for by the British government, and later it was mandated that a lord pay for the poor aid for the residents of his own estate. Which do you think is more despicable: spending taxpayer money to keep the Irish alive, or instituting a government mandate? Or should we have just let them all die?
Can you please introduce me to this person who does not want to have insurance? I am very curious to meet this person, because it seems to me that everyone talking about how we shouldn’t mandate insurance, and how being forced to buy insurance is a burden is insured. My absolute favorite was the anti-health care tea party representative who went on the record angry because his Congressional health insurance would not begin until February 1. As best as I can understand it, that is taxpayer funded health insurance. He might not see the hypocrisy, and maybe there isn’t any, but it seems quite heartless to me that he went on the record complaining about it.
You state: There’s another side to compassion, and that’s a willingness to admit that your adversaries might occasionally have legitimate or even lofty motives, and that their positions are sometimes worthy to be engaged, not simply pilloried.
Where in this do you admit that there is another side to compassion? You may STATE that there is another side, but then you go on and call the other side, to paraphrase, stupid and blind. That does not seem to me to be a willingness “to admit that your adversaries might occasionally have legitimate or even lofty motives.” In fact, your motives seem to be only to incite anger (although I admit that I was goaded in).
Are you really going to argue that the man who burst out “let him die” at the tea party rally was COMPASSIONATE?????
Finally, I’m sure that if you respond to me, your response will be something about how my thinking is not broad enough, but I think that’s rather a cop-out actually. Unfortunately, with the political realities of 21st century America it seems impossible that the United States will be funding healthcare for the world in the near future. To argue that we shouldn’t provide healthcare where we politically can because we are inconsistent and don’t provide it everywhere might be logical, but it seems like a logical cop-out to me. Why not fix the problems that you can fix instead of trying to note the inconsistencies in the problems that you can’t? Any response anyone will ever give you could always be criticized for being not quite broad enough, but sometimes thinking about every single permutation of the entire world is only possible in the ivory tower that is academia. Ultimately, some limit must be set or nothing will ever be done because it will never be good enough. But if nothing is ever good enough, is it not even worth it to try?
Jake:
compassion is a feeling for AN INDIVIDUAL, not for the nameless, faceless world at large. If you want to argue that letting the man die is better for the world at large, fine, but I object to your argument that this is compassion as most people would define it. Make up a new word if you must, but compassion is an EMOTION, and emotions do not always follow the law of logic.
But when we set government policies — when we decide whether to allocate that extra dollar to emergency medical care or to food banks — both the recipients of the medical care and the recipients of the food assistance are *equally* nameless, faceless, and anonymous. How, then, can compassion — by *your* definition — dictate one choice over the other?
Can you please introduce me to this person who does not want to have insurance?
I travel in circles where most people are pretty financially comfortable. I am guessing that in other circles, there are lots and lots of people who would prioritize many other things over emergency health insurance. I could of course be wrong. But can you please introduce me to the person who does not want a decent education for his children? Or heat in winter or food on the table?
Jake:
To start off with, no one is saying that the guy who yelled out “Yeah!” is compassionate. He sounds like a rather unpleasant fellow.
Many people (Paul Krugman, et al) believe that those of us who oppose government-run healthcare are slight variations on the Tea Party guy at the debate and, thus, not compassionate. This is not the case and Prof. Landsburg is explaining how.
All of your questions have answers (there are other posts on this subject in this blog). In a nutshell, resources are scare and there is a finite amount of money we can give to the poor and it is our duty to spend that money as wisely as possible. The poor need help paying for food, clothing, healthcare, car repair, Christmas presents, etc. Part of being compassionate is seeing what they really need or want before giving it to them. (SL tends to be more eloquent than me, so read what he has to say.)
Also, the vitriol that falls upon Krugman is generally well-deserved.
One pillar of your argument seems to be “If it were good to save him, he would have provided for his own healthcare. The fact that he didn’t shows that his priorities are elsewhere”
This may be true if, in fact, the guy was living in an environment where economically efficient transactions is a norm. However, that is emphatically not the environment he finds himself in. It is wrong. More than that, it is fractally wrong : http://www.flickr.com/photos/themadlolscientist/2421967468/
1) The fact that the US relies on a health insurance market for a big chunk of healthcare provision is already a big red flag. “The market for lemons” and all that.
2) You’ve already decried distortions (such as tax cuts to employers who provide health insurance) as making healthcare more expensive that it “should” be.
3) You’ve elsewhere pointed out that economic efficiency breaks down as a useful tool in exactly the kind of situation faced by this man – he may not have spent money on healthcare, but how much would you have to pay him before he’d be willing to give it up? Remember the parable of Bill Gates and bag of water in the desert?
4) The man made his choice under the reasonable expectation that he would be cared for in emergencies anyway. If you pull the rug out from under him now, it doesn’t make you compassionate.
5) My health benefits me, but it has positive externalities also. Therefore money should be spent to keep me healthy, even if I would not be willing to spend that much myself.
5½) The man made his choice when he was well. It doesn’t follow that he agrees with them now that he is sick. There’s a philosophical argument hiding behind this one, so I only count it as half a point.
That’s just off the top of my head. There may well be other ways that the “efficiency” argument breaks down.
Libertarian economists should not bother to talk about healthcare. The reality fails to fit their worldview in so many ways.
@Jake: You’ve lost the plot, mate. The critical statement you seem to have missed is when SL describes the more “heightened compassion” as: “[…]if you grasp and care about the underlying trade-offs”. Perhaps, it’s lost amongst the straw you were using to compose your comment.
The decisive point here is that when we consider how to alleviate the suffering of those individuals who invoke the emotion of compassion in us – by using government policy as a tool to that end – we should recognize that there are “trade-offs” to be made. In reality – as against your personal construction above – we cannot simply make an allocative choice to provide each individual with enough food to eat and enough medical insurance to guard against future medical costs – scarcity exists so we have to decide on the margin which allocation mix alleviates the most suffering. After all, that’s his point – being truly compassionate requires one to recognize this fact. To pronounce that you want everyone to have a full belly and medical care is nothing more than an amateurish, self-congratulatory notion that does nothing to actually alleviate the uneasiness of those that must decide between the two. The whole point is to alleviate the uneasiness as best as possible. To effectively do this – and, in point-of-fact, do what true compassion *demands*, SL poses the very relevant question – Shouldn’t we ask them which they prefer? Doesn’t asking questions like this get us closer to the effective alleviation of the uneasiness?
Wouldn’t public discourse and policy decisions – and, in the end, the recipients of support – be better served if people would recognize the need to make important trade-offs? And, instead of just shouting down those who they (perhaps wrongly) accuse of being devoid of compassion on the basis of whether they are willing to just blindly accept solutions based, not on reality (read: a world of scarcity), but on some imaginary construction where there are no values that cannot be satisfied by the correct allocation – wouldn’t we have better outcomes if they – people as smart as Krugman – would just admit that there are “deep and difficult trade-offs”?
That’s all SL is saying here. To go into tea-party politics and random, provocative shout-outs during debates is to FUBAR the point, entirely.
Harold: you did say the man didn’t want insurance. Nothing you have said indicates he did not want insurance, just that he wanted other things more.
When I say he doesn’t want insurance, I mean of course that he doesn’t want insurance given the alternatives available at the same price. Of course he’d want it if it were free.
When I say that I don’t want a Bentley, I think most people understand that I mean I don’t want to buy a Bentley *given its cost* not that I wouldn’t want one for free.
PS Harold: I appear to have hit a wrong button and inadvertently deleted your comment to which this is a reply! If you have a copy, I hope you’ll repost it! MANY apologies.
6½) Another point that needs to be kept in mind is that the US government is in fact not doing what voters say they want it to do – provide more, affordable healthcare, spend on infrastructure and job creation, and not bother about debt and deficit for now.
So you can’t dodge a “you’re compassionless” bullet by saying “if we asked people what they want, would they really want this money to be spent on healthcare?” The answer, after all, is “yes, they did.”
Doesn’t the citizen who chooses not to buy health insurance know full well that he will get Emergency health care, and as Landsburg says, thereby receiving emergency health insurance. So his preferences state that he’d rather have more money for groceries AND emergency health insurance as opposed to less money for groceries and emergency health insurance. There is no OR in this equation.
I meant to say “adverse selection” and not “moral hazard.” basically, my point is that at least when you force poor people to buy health insurance, it appears to me you will likely provide them with a deal they otherwise wouldn’t be able to get if most were not forced to buy. This is different from forcing them to buy groceries, where the concept of adverse selection isn’t relevant. In other words, it seems like if you’re going to force a group to buy something, make them buy insurance.
What is seen and what is unseen?
Anybody can see the poor lad arriving in the ward and demanding help and a lot of compassion.
The question is raised, is it worth this amount of compassion?
Maybe, maybe not. Whether you look at it as an economist, is imho not very relevant.
However, as an economist, you should also look at the unseen.
The unseen would perhaps reveal that in fact a lot more of the compassion is addressed to the poverty stricken medical establishment, enjoying massive privileges in their trade (big artificial restrictions on the supply of medical staff) and extracting huge compassion rents out of the rest of society.
As an economist, you could therefore also ask yourself, is this trade worth this much of compassion?
Maybe, maybe not.
Since there are a number of developed countries with much less compassion for medical cartels, an empirical economist could even dig up the data and compare notes.
An economist could also notice that the two kinds of compassion are highy correlated: the more compassion is extracted from society for the latter, the more compassion will be needed to help the former.
Who knows, perhaps an obscure economist might be found who dares suggest that a free market be imposed in the medical trade, possibly raising the whole society’s wealth.
Unfortunately, most blogging economists prefer to discuss and treat symptoms, not causes.
“To pronounce that you want everyone to have a full belly and medical care is nothing more than an amateurish, self-congratulatory notion that does nothing to actually alleviate the uneasiness of those that must decide between the two.”
Putting aside for a moment the issue of suffering abroad, there is no reason to choose between the two. The USA is really quite wealthy enough to provide basic medical care and food for everyone. The question is why should we provide food and medical care instead of, say food and cinema tickets, or cinema tickets and beer vouchers?
Steve’s point is that we should ask the recipients where they want it spending. Most people seem to buy food, so we know they value food. Some people do not buy healthcare, so they do not value health. If we had to choose between food and healthcare, then on the basis of these revealed preferences, we should give food rather than healthcare. We should not let one person starve because we spent the money of someone elses healthcare.
However, as I said above, we have enough to provide both. Say we have spent enough to feed everyone, and have money left over to provide healthcare for all. Should we actually spend it on healthcare, or on something else? We have already said that we should provide enough for food and some extra, what if their choice was for food and shelter, rather than food and healthcare? We could spend it on shelter and not healthcare, or we could say, no, lets put enough in to provide food, shelter AND healthcare. After all, no one in such a wealthy country should be without shelter.
We can carry on, providing the basics, looking at peoples choices, and providing enough for everyone to have whatever it is just before the healthcare. At some point we must stop, if peoples choices get to the point where they prefer something we do not believe should be a minimum standard in a civilised society. So if people prefer cars to healthcare, we may say that it is not really a requirement for society to provide cars, so we shouldn’t provide healthcare either.
Another way is to put in as much money as you think is desirable, and keep funding up the choices until the money runs out. If it runs out before you get to healthcare, then you let the people die.
As I said above, I do not believe that peoples utility is necessarily maximised by their own choices, particularly when they are in positions of duress caused by destitution. It may be appropriate to construct our heirarchy based on criteria other than revealed choices.
“It was specified here that this was a guy who didn’t want health insurance.”
Mr. Landsberg, does this hold true if his economic realities forced him into a long-shot bet? If you can’t feed yourself THIS WEEK, then is he truly stating he didn’t want insurance, or the fact that he was hoping to effectively win the lottery? (The ‘lottery’ here being that you and all your loved ones never need emergency catastrophic health care and all die peacefully in their sleep at an advanced age)
I feel there is some sort of distinction between truly ‘doesn’t want’ and forced to bet he doesn’t need. His situation almost seems to remove him from the rational economic universe. He can’t starve so he has to just hope and pray he won’t suffer an accident…which…in fact…is quite rational…since if he starves this week he won’t need health care in the future. But I’m not sure this truly moves him into a ‘doesn’t want’ category.
I always enjoy how much emotion and response this subject brings. It appears a few posters have forgotten that resources are scarce (thanks to BZ for pointing this out). A country can control its purse by increasing taxes – and maybe that is the answer – but ignoring that there is a trade off to be made that involves more than the individual in need is ignorant.
A key point that I feel is being ignored is that individuals are making a decision to consume something over other goods/services/actions. Individuals in America do have some control of their income level (assuming the rational individual, omitting those without the capacity to make rational decisions, which is another discussion). I admit that an individual is bound by constraints in their ability to advance their economic status however, there are choices involved with being at the margin and these choices have an economic impact.
The choice to consume more leisure, to increase family size and how much effort to put into gainful employment (including acquiring the signals needed to obtain employment such as education, tidiness, hygiene, etc.) has both current and future costs. I believe that access to education (K-12 at the least); methods to control family size, scholarship and the like are widely available to the population as a whole. The argument that preferences change and the person at the emergency room should receive care based on a change in preferences, thereby maximizing current utility ignores lifetime utility maximization.
Building off of previously established principles that individuals make decisions taking future income into consideration, one can logically expand that to individual decisions in consumption of goods and services. This is a similar debate to the one of smoking and providing care to those with cancer from smoking. Smoking has a huge societal cost through loss of life and health care costs. Everyone that has health insurance is paying higher premiums because people smoke, yet we still allow people to choose to smoke. By not doing so we are infringing on their freedom to choose. There is no moral or ethical examination of this, no pause for determination of right and wrong; it is a sterile observation, which is a large part of what good economic thought should do.
Given that individuals have a choice as to how to distribute their time, energy and income; and that they have some level of control over the level of their income, it follows that there is some level of individual responsibility that must be assigned/taken for the choices made. Additionally, understanding that the amount of public spending by the population (through direct gov’t spending and subsidies through pricing policies) is limited by the populations’ income and willingness to be taxed; difficult decisions must be made. The crux of the argument is that to truly maximize the nations (and individual) utility, we must both consider the lifetime maximization of utility at the individual level (while assigning responsibility where it belongs) and remove emotion evoking terms such as “compassion” and “deserve” from the debate as they do not fit into the decision matrix.
What’s seemingly lost in the discussion was the stipulation that the sick person “has a good job, makes a good living” and turned down 200 to 300 dollar a month health insurance. That’s not quite the picture that everyone was painting here of the poverty stricken choosing between food and health insurance. http://archives.cnn.com/TRANSCRIPTS/1109/12/se.06.html
Unfortunately there was little discussion of how one cares for the guy at the emergency room, because many of the so called compassionate ways to care for him are inefficient and would result in a far greater trade-off with overall medical resources or resources in general.
It’s one thing to provide a kind of health care voucher or universal catastrophic coverage program, it’s another to force everyone into a highly regulated and cartelized system with community rating and guaranteed issue.
The manner in which government intervenes to address individual cases can reduce the standard of living of society at large, but we feel better about it because no one cries about things that were never created.
Overall this is an interesting question – and one that Steve brings up in his books.
I once was listening to Dr. Joy Brown on the radio describe her trip to Russia before 1992 (If I remember right). She described her train ride to her destination. The whole train was lavishly decorated – a train ride fit for a King or Queen. She recalled that she was absolutely enamored with it because of the ideals that were behind it. The idea that every person should travel in luxury – not just the rich. She remembered how such a powerful message moved her.
Of course, she’s a Psychologist and not an Economist, so I don’t think it ever occurred to her that the money put into making such a beautiful train might have been spent instead on feeding the people of the nation.
Nowhere in this discussion have I seen any mention of an enforcement mechanism. Who will be in the emergency room with the authority to say “no treatment for you”? A doctor? We give physicians and other medical practitioners extraordinary privileges (how many of us can cut people open with a knife without going to jail?) and expect extraordinary behavior in return. Basically, a doctor, upon entering the profession, is on call to perform life saving actions all the time for the rest of his or her working life. This code of conduct is so much a part of the self-image of the profession that it is almost unthinkable that doctors would agree to undertake the odious task of shunting the uninsured aside to die. Would you trust a member of your family to a doctor capable of such cold-bloodedness? So who would play the grim reaper? Cops? Social workers? Economists? Even in a down economy, I don’t think there would be many takers for these jobs, and you might not want to meet any of them at the door of the emergency room, even with your insurance card in hand.
Oops! One more thing:
I don’t know if I’m too far off here, but I think we can compare the use of the word “compassion” to the use of the word “taste.” Once you start using these things to describe behavior, then you’ve just divided by zero – anything can therefor be explained away by such a bland statement.
The whole ‘resources are scarce’ argument when applied to the US is getting a bit old. It’s rather hard to convince proponents of UHC that distribution of resources are an obstacle when nearly every other industrialized nation has a form of UHC, as well as some developing nations that are ranked higher than the US.
Has anyone considered that a healthier pool of workers would make them more productive, thus increase real wealth in the economy?
I have health insurance, in fact, very good health insurance. However, my coverage has a lifetime cap. So, the situation of the young man showing up at the emergency room is not in principle different from what would happen if I required some very expensive experimental treatment to stay alive. If my treatment would exceed the lifetime cap on my policy, my insurer would refuse to cover the treatment. Is the hospital obligated to treat me anyway? I doubt it.
In the end, in the U.S., we have decided to discriminate in access to health care based on ability to pay. All countries do this, just to a greater or lesser extent. In U.K. citizens all have access to the National Health, but if they can afford it they can get treatment through private channels that is not covered under the NH.
In the end, this is a debate about the level of discrimination (used not in the perjoritive sense, but in the differentiation sense) we want to apply to health care – to what extent do we want to discriminate based on ability to pay.
Admitting that resources are scarce is not an argument against UHC, it is a statement of fact; trade-offs exist. The public cannot be expected to finance the satisfaction of every taste. I use the term public as a combination of gov’t spending, charitable contributions and any other wealth transfer, voluntary or involuntary, not just gov’t program spending. A number of the nations are in dire financial shape partly due to paternalistic programs.
That said, I agree that a UHC system may be a positive sum proposition in some form however an argument that ‘other countries have it, so we should be able to afford it’ isn’t very solid ground to stand on. A lot of people in my building have a luxury car, that doesn’t mean I should get one or should be able to afford one, etc. This is flawed logic. The decisions on how to allocate resources should be based on the best use of those resources according to the consumers, not because someone else has it, thinks it is best or wants the consumers to have it.
Personally, I prescribe to Sen’s view of economic development; which accounts increases in health as utility producing. Again, there may be merit in UHC – I don’t believe that anyone has gotten the math right accounting for all of the variables that will allow for a good decision to be made. Policy decisions should be based on outcomes driven by economic development and not some emotional reward (which has a pseudo-quantifiable property as well, BTW) which should allow for better policy thereby helping more folks than if we target pet programs without regard and understanding of the cost/benefit tradeoff just because it ‘feels wrong’ not to do so.
@Bill, if you view health as utility producing, then I assume you would agree that our health priorities in the U.S. are very skewed. From a utility or contribution perspective, I would argue that we should provide UHC to all minors, and then have UHC in a decreasing subsidy as we age, providing no coverage for people once they retire – quite the opposite of our current system.
@Al V.
The example of your lifetime coverage cap is not actually valid, and you might want to go and check your health care coverage again. Under the PPACA, lifetime caps on coverage have been illegal since September 2010.
@Al V, you know what they say about assuming. ;) My original point is that policy paternalism based on emotion isn’t the best approach. What is your reasoning for you prescription of UHC to minors and not adults on a regressive scale?
Steve,
My history of arguing this issue tells me that your approach is a loser. You might be technically right, and it might be a good thing to get people thinking of the tough choices involved, but, for health care, most people have a complete double standard.
This has led me to advocate a two-tiered system. We would have a baseline system a lot like Canada’s. It would suck, and folks would die waiting to be seen or receive treatment. You couldn’t sue anyone and you would deal with classic rationing in order to meet cost goals.
In addition to the first tier, the second tier would include the classic elements of our current world-beating health care system. I would imagine that the first tier would be good enough for a large share of health care — colds, flu, cuts, sprains, etc. Even folks with access to the second tier would go to the first tier for those simple things. More exotic diseases and responses would probably land those that desired to pay into the second tier of treatment.
The beauty of this system is that nobody has be “left to die”. No, they go right into the system. Nobody with cancer in Canada is considered to be “left to die”. Even though with a disease like cancer time is of the essence and the delays of the system mean a much greater mortality rate.
This way we can all be happier. We cut down on our health care costs, and nobody has to feel bad about it because we aren’t “letting people die”. (We are just putting them in line.)
I don’t know much about economics, but I have heard the slogan “rational people think at the margin.” If that’s the case, then shouldn’t we act as follows:
1. See a poor person in need of immediate help, like dying in the street or staving
2. Evaluate the marginal benefit of helping him (i.e. saving his life)
3. Evaluate the marginal cost of helping him (i.e. spending resources which could have been used in future to do something else)
4. Weigh the two
Considering diminishing marginal utility, doesn’t the fact that we have so much resources as a society imply that the marginal cost of helping him is probably really low in the grand scheme of things, because we’ll still have plenty of resources to help others?
And assuming that the marginal benefit of helping him is pretty high, and that we’re those “rational people” that economics talks about, wouldn’t we help him?
Keshav Srinivasan: There are at least two problems with your program.
1) The “marginal benefit” of saving a person in dire straits is not well-defined; one could interpret either as the amount he’d pay for help, or the amount we’d have to pay him to forgo help.
2) One has to decide whether to weigh marginal costs and benefits ex ante (i.e. before we know who will get sick) or ex post (i.e. after that uncertainty has been resolved). You are arguing for making an ex post calculation, but there are powerful arguments for making these calculations ex ante. (I’ve made some of those arguments in my books.
One advantage of calculating ex ante is that, as long as any individual’s probability of falling sick is small, the two definitions of “marginal benefit” will (approximately) coincide (this needs some assumptions and takes some proof, but the assumptions are reasonable and the proof is valid).
You have to be careful about this stuff. Suppose, for example, that we adopt a policy of ex post evaluation, together with a “willingness to accept” definition of marginal benefit. Then when you get sick, you might not be willing to accept any amount at all to forgo treatment; it follows that we are obligated to devote all the world’s resources to saving you (if that’s what it takes). But ex ante, nobody would want to live in a world where the first guy to get sick claims 100% of everyone’s resources.
My argument in the post is precisely that we should be calculating ex ante. If we’re out to help poor people, we should ask them “Would you rather live in a world where you’re eligible for emergency medical care, or would you rather live in a world where we spend those same resources to give you a better education?”. Once you’ve done the ex ante calculation and found the right policy, you can’t also commit to making ex post calculations that undermine that policy.
Steve, I’m not taking his interests into account at all. I don’t care how much he values what we’re doing for him, whether we’re talking about how much he’d pay for it or how much compensation he’d want to give it up. When I was talking about marginal benefit, I was talking about the moral value of saving his life, versus the moral value of being unable to help or save the lives of others because we have no resources left.
If we think about it from that perspective, we have to judge how each deed we do affects the moral standing of society. When we let he poor die in the street, we are arguably morally responsible if we were able but unwilling to do anything. And we are equally responsible for the lives saved if we give up our resources to help those in need. Again, I return to diminishing marginal utility. When we have so much resources to fulfill our moral obligations in future, and just a little bit of those resources can do great good right now, it’s a no-brainer.
I can understand why an individual would just seek to maximize what he values most, but shouldn’t we collectively remedy our individual moral failings and focus on what *ought* to be done, not, or at least not just, on what people want?
“2) One has to decide whether to weigh marginal costs and benefits ex ante (i.e. before we know who will get sick) or ex post (i.e. after that uncertainty has been resolved). You are arguing for making an ex post calculation, but there are powerful arguments for making these calculations ex ante. (I’ve made some of those arguments in my books.”
IMO, this is not a valid statement. Technically, it is–but only if there was a valid need to do so. There is not. The need/demand for healthcare is a standard and well-known distribution, so the need to track the specific choices of each individual is irrelevant (and adds to the cost). The only need is to cover the actual costs incurred by the system–which will vary year to year–but will be reasonably predictable a couple years in advance. Thus, a UHC system would save money and time for everyone–because the real issue with healthcare is the *unpredictability* of need by any one average individual. As a UHC system means a 40% to 50+% reduction in overall medical care costs (if PPP per capita costs drop to the same as France), that frees up resources to be spent in other areas. Federal expenses fall significantly (Medicare), state expenses fall as well (Medicaid savings split between state and feds), and then the consumer saves as well (lower medical costs AND lower govt costs). Estimated annual savings are $1T for UHC to $1.5+T for SP/UHC.
Keshav Srinivasan: I am moderately well convinced that most of the time, the moral standing of society is best enhanced by equating marginal costs with marginal benefits, as measured by those who bear those costs and receive those benefits. Again, I’ve written about this at some length in ##tbq, though I realize there are still a few people who either haven’t read those chapters (!) or were unconvinced by them.
Jerry: You are addressing a different question than the one I was trying to raise. This isn’t an issue of predictability. Even in a completely predictable world, the same issue would arise.
E.g.: Suppose I currently face a 1-in-a-million chance of needing emergency care (this probability being known for certain, so that exactly one out of every million people like me will need this care). I am willing to pay $10 for the assurance that I’ll get this care if needed, and so are a million others like me. On the other hand, after one of us gets sick, he’s willing to pay $5 million for that same care.
Should we calculate the benefit of that care as $10 million (that is $10 per person times 1 million people) or as $5 million? That’s the difference between an ex ante and an ex post calculation, even in a world where we know for certain that exactly one person gets sick.
“as long as any individual’s probability of falling sick is small, the two definitions of “marginal benefit” will (approximately) coincide”
The chances of an individual falling sick is very large over a lifetime.
His consumption choices will only reveal his preferences IF they affect his consumption.
Since in America he will get emergency health care no matter what he spends his money on, his spending will not actually reveal what his preferences are.
Indeed the rational choice for someone who ONLY wants emergency health insurance is to claim they don’t want ANY health insurance. Then they get it for free.
Ben:
His consumption choices will only reveal his preferences IF they affect his consumption.
Since in America he will get emergency health care no matter what he spends his money on, his spending will not actually reveal what his preferences are.
Excellent point. Thanks for this.
@Bill, my thinking is that it makes sense to invest in children to facilitate their productivity as adults. Children who have poor medical care miss more school, and their conditions often lead to degraded productivity as adults. I would argue that giving them access to to medical care improves the likelihood that they will be productive adults, and my guess is that a relatively small investment in childhood medical care would have long term benefits.
On the flip side (and this is coming from someone approaching retirement), retirees are basically a drain on society. They don’t work and they cost us a great deal of money in retirement benefits and medical expenses. Better to expedite their demise; or perhaps simply to limit the amount we spend on their health care.
Our socialist country runs a vast network of national parks, forests and monuments that the poor, Blacks and Hispanics don’t need, don’t want and never visit.
The nanny-state answer is that “there’s nothing keeping them from visiting.”
Obamneycare is not health care. No health insurance, including Medicare, does anything for the guy climbing Everest, solo sailing around the world or simply living retired in India, Thailand, Brazil or even France, where great health care is available to anyone who hasn’t spent 45 years wasting his wealth paying into stupidities like health insurance and Medicare.
Steve, for the record I did read the Big Questions (and the Armchair Economist), it’s just that I’m a deontologist, not a consequentialist. I’m fine if social welfare is less than optimal, as long as the moral state of society can be made higher. But I guess you could take people like me into account in the Economist’s Golden Rule. You can say that the utility of some people in society, let’s call them altruists, depend on what happens to other people, not necessarily on the other peoples’ utility, but on their “well-being” as judged by the altruists. I guess you would model this by a nonlinear system of differential equations, so it may be mathematically intractable, but in principle it could be done.
“E.g.: Suppose I currently face a 1-in-a-million chance of needing emergency care (this probability being known for certain, so that exactly one out of every million people like me will need this care). I am willing to pay $10 for the assurance that I’ll get this care if needed, and so are a million others like me. On the other hand, after one of us gets sick, he’s willing to pay $5 million for that same care.”
Given your premise of ONLY 1:1-million incidents per year, then there is no need for any more of that treatment needed after the one case occurred (“perfect knowledge”). Your own example violates your own premise of 1:1-million, because you claim there will be a second case when your premise is only *one* case will occur.
Steve, I think you are giving these people way too much credit. The “let them die” crowd is composed or people who either get hard sticking it to “big government” or don’t care about poor people.
Let me give a related example of an issue where liberals act similarly. I talked to a girl about the death penalty. I said the death penalty might save more lives through deterence, and since the empirical evidence is so murky I’m sticking with my prior that it does. She said that she doesn’t care what the tradeoff is, she thinks its horrible. (This is probably what that crowd would say.) So I asked if she’d execute a pedophile to save a woman and child if all of these hypotheticals were known facts. She said no.
She’s was just too big of a coward and she’d rather let them die than man up. I think that’s how these people in the crowd felt. This was about sticking it to Obama(care) and if you brought up the tradeoffs, making the case to subsidize care and save the guy, it’d make their head spin and they’d say “f*ck ’em and Obama!”
But I’m glad you have more faith in people than I do.
Jerry: I’m not trying to give you a realistic example; I’m trying to give an example that illustrates the ex ante/ex post distinction without cluttering it up with a lot of realistic but irrelevant details.
We have a population of 1 million people. We know that one year from now exactly one will be stricken with a horrible disease. We know that each of the million would be willing to pay $10 upfront to know that a cure will be available if they get sick. We know that whoever is stricken will, upon being stricken, be willing to pay $5 million for a cure.
We can develop a cure at a cost of $7 million.
To what benefit should we compare that cost: The ex ante benefit of $10 million ($10 times 1 million people) or the ex post benefit of $5 million?
This is fundamentally the same issue that arises with emergency room care in a complicated world full of uncertainty. I’m posing the question in a less complicated model in order to focus attention on what’s essential.
There are good arguments for doing ex ante calculations. In the real world, those arguments translate into weighing costs against what people are willing to pay before they get sick, not what they’re willing to pay (or willing to accept) after that bit of uncertainty is resolved.
Keshav Srinivasan: As far as accounting for the philosophical preferences of people like you:
1) I once wrote a paper on this subject.
2) Ted Bergstrom had a remarkable article in the American Economic Review a few years ago where he gave a very surprsing (to me) argument against counting altruistic preferences in the social welfare function. I’ve been meaning to blog about this for some time, and might soon.
“To what benefit should we compare that cost: The ex ante benefit of $10 million ($10 times 1 million people) or the ex post benefit of $5 million?”
Neither–because the real cost is not a “point in time” one-time benefit. Cost needs to be amortized over the reasonable expected lifetime of the benefit (once developed, it does not need to be developed again).
The claim could be made a different *unique* illness will strike one person and need a unique different tratment each year–which will have that same type of cost to treat. However, being willing to spend $X on a one-shot treatment will be patient-specific, and the amount will be based on their present and future expected income streams (and whether they can actually *pay* the money for the treatment). The “uninsured” person is willing to spend $X to treat him/her self if they *lose* their bet. That is their personal perceived risk cost of choosing to not be insured. If the person is *not* willing to pay the actual cost of treatment if they lose that bet, then he/she pays the penalty (death/disability/whatever).
@ AL V – That’s one way to address the issue with social security I suppose. Your point is well taken and serves to reinforce my previous point that health can increase utility at both the micro and macro level. The slippery slope is how much public intervention should be provided. For example, it could be argued that immunizations provide a positive value for subsidization or should be a public service. Especially for communicable diseases much like some of the programs that exist for IV drug users. Uniformly address the health issue at is source for a per unit fee and reduce or eliminate the geometrically higher costs and uncertainty in the future. Beyond this I find it difficult to argue favoring any group over another based on demographic information or any other factor, there should be incentives and information in the market to allow for *near* optimization.
“I said the death penalty might save more lives through deterence, and since the empirical evidence is so murky I’m sticking with my prior that it does. She said that she doesn’t care what the tradeoff is, she thinks its horrible. (This is probably what that crowd would say.) So I asked if she’d execute a pedophile to save a woman and child if all of these hypotheticals were known facts. She said no.”
I assume you are refering to the “8 lives per execution” conclusion from Ehrlich. Well, one of the points was that the deterence effect was actually very minor compared with other social influences, such as poverty, and small compared to the chance of conviction. I assume you would agree that if you want to prevent murders you should address the more important factors first? Executing one murderer MAY save 8 future victims, but reducing unemployment may save far more, and better policing more still. Remember, 8 is a very small proportion of murder victims.
Harold:
I assume you are refering to the “8 lives per execution” conclusion from Ehrlich. Well, one of the points was that the deterence effect was actually very minor compared with other social influences, such as poverty, and small compared to the chance of conviction.
One might add that Ehrlich has more recently updated this estimate to something as high as 24 lives per execution, and that he remains a staunch opponent of the death penalty.
Late to the party again but wanted to chime in on “the man who can’t be treated”.
Jonathan Swift also said “every man desires to live long, but no man would be old.” At some point we risk flattering Mssr. Krugman with criticism, but “freedom to die” really is a gigantically lame straw man. So a potentially interesting hypothetical about an adult who specifically CHOSE not to purchase coverage somehow morphs into a poor child (an example even most libertarians would view as categorically different). A simple extension of SCHIP would’ve sailed through (as I recall some questioned whether it should be expanded to households making up to $80,000), and the “eruption” of cheers was probably about the same as if a fringe Democrat had been asked if Bush planned the 9/11 attacks.
On the economics, the recurring idea that we don’t need to bother with trade-offs because the US has ‘extra’ resources also seems to reflect a lack of “imagination”, e.g. a fuzzy/static notion of things like “healthcare” as widgets. How can one say “we can do it all” when nobody knows what “it” will/could be, what cure we might find next? How about more education instead? Or better education? Or even better education? You know, to “win the global economy” (actually once upon a time Mr. Krugman actually wrote quite eloquently about how this view of countries competing is bogus). I presume what some people are implicitly referring to here is a set of “basic” public HC services, and most of us might even agree with that conceptually. But let’s not pretend that at that point (e.g. saying “France is #1”) we’re not drawing all kinds of gut-wrenching and seemingly arbitrary lines. Krugman likes to use the phrase “the care we need”, which is not very helpful. And since today’s cutting-edge research becomes tomorrow’s basic care, a compassionate observer might consider the impact of our choices on future lives (possibly many more lives?) from any resulting disincentives to innovate. The very (tempting) idea that we can have the “best of both worlds” by expanding the “to do” list, without transforming the system that produced the ‘extra’ resources in the first instance, seems illusory if we can expropriate ‘extra’ resources only once.
So to me the real problem with Blitzer’s hypothetical is that it’s not specific enough. How can Mr. Paul or anyone really answer the question thoughtfully without knowing WHY the hypo man needs 6 months of intensive care? (And yes, at what cost as well.) Our desire to avoid these difficult choices is why Medicare and Medicaid are incurably insolvent, and it seems an important part of the motive for UHC for many people has less to do with alleged economic efficiency than the redistributive aspects of (coercively) broadening the pool. The fact that some people might have access to HC services that others don’t (through any responsibly-designed public option) may be politically unpalatable, but at the end of the day the alternative is to prevent some people from accessing services they would have been willing (and able) to pay for privately, rather than have an honest dialogue about the nature and extent of our moral obligation to provide assistance to those who need it.
P.S. The other main trade-off-avoidance argument, that preventive care saves money, seems like an empirical issue that is usually just asserted. Surely there are areas where an ounce of prevention is prudent, e.g. to guard against true “public health” risks like contagious disease (which I presume is why we have a CDC). But as a general rule, more preventive care would be expected to save money in instances where a preventable condition is detected, otherwise it just represents additional spending.
And as for the idea that not having UHC is undermining our economic productivity relative to other nations…c’mon.
Sorry that got a little long.